Although death certificate data are limited by possible misdiagnosis, incomplete case ascertainment, errors in coding, etc., detailed mortality information on some neurologic diseases for the entire U.S. is not available. Analysis of mortality data can be particularly useful for some neurologic diseases because these may indirectly contribute to death. Since there are no uniform criteria for what constitutes the underlying cause of death in patients, it is important to examine all deaths in which a disease is listed as an underlying, immediate, associated, or contributory cause of death to get more complete information about the relationship between the disease and death. Association of diseases occurring at the time of death are also being studied for all deaths occurring in the U.S. for many neurologic diseases. Diseases occurring together may provide important information in the search for etiology. Detailed analyses of mortality data have been accomplished for Huntington's disease (comorbidity at time of death), Friedreich's disease (survival analysis), international comparisons of average annual age-adjusted mortality rates for primary tumors of the nervous system, and stroke (case-control study of risk factors).